VLSplat: Vision-Language Guided Object-Centric 3D Gaussian Splatting via Scene Graph
cs.CV, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Project page: https://vlsplat.github.io
Terminology
Sources
- Gaga: Group Any Gaussians via 3D-aware Memory Bank
- Learning Segmented 3D Gaussians via Efficient Feature Unprojection for Zero-shot Neural Scene Segmentation
- Semantic Gaussians: Open-Vocabulary Scene Understanding with 3D Gaussian Splatting
- GLS: Geometry-aware 3D Language Gaussian Splatting
- OVGaussian: Generalizable 3D Gaussian Segmentation with Open Vocabularies
- SAM 2: Segment Anything in Images and Videos
- The Replica Dataset: A Digital Replica of Indoor Spaces
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs